Thanks for the info, much appreciated. However this kinda makes things more confusing... I mean, if the optimal values for master/kernel parameters depend on the actual program (particular sampling process) and not on the type of effect (general characteristics), why are these parameters global and open to the user instead of per-preset and left for the developer to set up?
Also, when you say kernel length what parameter exactly are you referring to: the number of kernels in Master page, the different parameters in Kernel page, anything else?
In any case, I understand from what you say that the default prameters for the Nebula and NebulaReverb dlls are basically "optimized" in general terms and we shouldn't touch them... so I assume that we should only set LFreqD and/or RateCNV for long reverbs when their tail is cut out and that's it. Is this correct?
In any case, I was just looking for precise, "general optimisation" tips for each of the effect types listed above. Would I be correct in assuming that I should use the nebulareverb dll config for reverbs and the plain nebula config for the rest of the types?
Oh and thanks again for the input, any info from experimented developers on these obscure issues is always more than welcome.